Differentiating Between Pancreatic Adenosquamous Carcinoma and Pancreatic Ductal Adenocarcinoma Using Computed Tomography
Canadian Association of Radiologists Journal · 2026
DOI: 10.1177/08465371261426959 ↗Kurz gefasst
Fachartikel in der Zeitschrift Canadian Association of Radiologists Journal, 2026. Verfasst von Marwa Alkatheeri und 11 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, Hôpital Cochin, AP-HP, Paris, France. Hauptthemen laut MEDLINE: Carcinoma, Adenosquamous, Tomography, X-Ray Computed, Pancreatic Neoplasms, Carcinoma, Pancreatic Ductal. Die Arbeit nennt 37 Literaturangaben.
Abstract (Originalsprache)
Purpose
To identify CT features that may help discriminate between pancreatic adenosquamous carcinoma (PASC) and pancreatic ductal adenocarcinoma (PDAC).
Methods
CT examinations of 42 patients with PASC were analyzed by 2 independent readers and compared to those obtained in 42 patients with PDAC. Sensitivity, specificity, and accuracy of each variable for diagnosing PASC versus PDAC were calculated. Associations between variables and PASC were examined using univariable and multivariable analyses. A CT signature was developed to distinguish PASC from PDAC.
Results
PASC presented as single (100%), oval (93%), and heterogeneously enhancing (98%) pancreatic mass with a mean largest diameter of 46.2 ± 18.5 mm, vessel encasement (69%), and segmental portal hypertension (64%), in association with hepatic metastases (52%). In univariable analysis, ring enhancement (odds ratio [OR], 25.23; P = .002), internal necrosis (OR, 9.48; P < .001), heterogeneous tumour enhancement (OR, 25.23; P = .002), and segmental portal hypertension (OR, 4.5; P = .001), were the most discriminating CT findings for the diagnosis of PASC. In multivariable analysis, internal necrosis (adjusted OR, 5.44; 95% confidence interval [CI]: 1.83-17.34; P = .003), heterogeneous tumour enhancement (adjusted OR, 8.80; 95% CI: 1.41-171.47; P = .049), and segmental portal hypertension (adjusted OR, 2.96; 95% CI: 1.02-8.97; P = .048), were independent variables strongly associated with the diagnosis of PASC. The CT signature yielded 81% sensitivity (95% CI: 66-91), 74% specificity (95% CI: 58-86), and area under the receiver operating characteristic curve of 0.840 (95% CI: 0.756-0.924) for diagnosing PASC.
Conclusion
CT examination demonstrates several features that help discriminate between PASC and PDAC. A CT signature based on 3 imaging features can help distinguish PASC from PDAC.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Radiology, Hôpital Cochin, AP-HP, Paris, France.
- Faculté de Médecine, Université Paris Cité, France.
- Department of Gastroenterology, Hôpital Cochin, AP-HP, Paris, France.
- Department of Pathology, Hôpital Cochin, AP-HP, Paris, France.
- Department of Hepatobiliary, Endocrine and Digestive Surgery, Hôpital Cochin, AP-HP, Paris, France.
ORCID-Profile: Maxime Barat, Philippe Soyer
Zitieren
APA
Alkatheeri, M., Barat, M., Moyne, T., Pellat, A., Dohan, A., Nguyen, T. T. L., Bonnin, A., Coriat, R., Dautry, R., Rouquette, A., Marchese, U., & Soyer, P. (2026). Differentiating Between Pancreatic Adenosquamous Carcinoma and Pancreatic Ductal Adenocarcinoma Using Computed Tomography. Canadian Association of Radiologists Journal. https://doi.org/10.1177/08465371261426959
Vancouver
Alkatheeri M, Barat M, Moyne T, Pellat A, Dohan A, Nguyen TTL, et al. Differentiating Between Pancreatic Adenosquamous Carcinoma and Pancreatic Ductal Adenocarcinoma Using Computed Tomography. Canadian Association of Radiologists Journal. 2026. doi:10.1177/08465371261426959
BibTeX
@article{Alkatheeri2026,
author = {Alkatheeri, Marwa and Barat, Maxime and Moyne, Thibault and Pellat, Anna and Dohan, Anthony and Nguyen, Thi Thuy Linh and Bonnin, Angèle and Coriat, Romain and Dautry, Raphael and Rouquette, Alexandre and Marchese, Ugo and Soyer, Philippe},
title = {Differentiating Between Pancreatic Adenosquamous Carcinoma and Pancreatic Ductal Adenocarcinoma Using Computed Tomography},
journal = {Canadian Association of Radiologists Journal},
year = {2026},
doi = {10.1177/08465371261426959}
}
EUVIMED